Yue Ma

Projects

What I build and what I contributed. Papers are on the publications page; full detail is in the CV.

Agentic systems and scientific evaluation

Three layers of the same problem: a system that carries out the research, a methodology for representing what any such system did, and a benchmark that asks the question inside one field.

  • 01 · Agentic system · Feb 2026 – present

    SIDERIUS

    A closed-loop agentic system for autonomous research: it proposes, implements, and evaluates machine-learning experiments under real compute budgets. Work spans multi-agent workflow design, online resource control, and reliable long-horizon execution.

    • Sole architect and developer

  • 02 · Methodology and infrastructure · Jul 2026 – present

    SciTra

    A methodology, and the infrastructure that implements it, for representing what a scientific agent actually did: how a decision trajectory is recorded, and how the action space is defined, so that runs from different agents, tasks, and fields can be read on the same terms. Resource use is one of the quantities that representation makes measurable. None of it is tied to a single benchmark or a single field. Built with 50+ scientists across 20+ institutions in collaboration with BenchFlow.

  • 03 · Domain benchmark · Jul 2026 – present

    FrontierPhysics

    A benchmark in one field: how AI agents carry out frontier physics research iteratively, from literature review through research-plan implementation, on tasks that take PhD-level researchers weeks. Built by the BenchFlow team with contributors across universities, national laboratories, and industry.

Machine learning research

  • Nov 2024 – Jun 2025

    Adaptive Neural Processes for Extreme Forecasting

    Probabilistic forecasting under distribution shift and limited observations, with a focus on few-shot adaptation to extreme regimes never seen during training.

  • Jul 2024 – Oct 2024

    Multi-Fidelity Surrogate Modeling for Scientific Simulation

    Learning efficient surrogates for computationally expensive, high-variance detector simulations, combining Gaussian Processes and Neural Processes for rare-event emulation.

    Contributed core algorithms and derivations to the ICLR 2025 Spotlight paper “RESuM” (acknowledged for significant contribution).

Statistical inference and scientific computing

Research within the XENONnT Collaboration, a 200+ scientist international experiment operating petabyte-scale data systems for dark-matter and solar-neutrino searches.

  • XENONnT · Oct 2020 – present

    Probabilistic modeling and statistical inference

    Bayesian and maximum-likelihood inference for the first measurement of solar pp neutrinos at ultra-low momentum transfer, including calibration, data selection, and Monte Carlo workflows for uncertainty and sensitivity estimation.

  • Maintainer · Jan 2024 – present

    strax and straxen

    Open-source data frameworks processing petabyte-scale detector data for 200+ collaborators, with work on ETL performance and memory-efficient processing.